CHADS₂ Score for Atrial Fibrillation Stroke Risk Assessment
The CHADS₂ score is a validated stroke risk stratification tool that assigns points for Congestive heart failure (1), Hypertension (1), Age ≥75 years (1), Diabetes (1), and prior Stroke/TIA (2), with scores ≥2 warranting anticoagulation with warfarin (INR 2-3) to reduce stroke risk and mortality. 1
Score Calculation and Components
The CHADS₂ scoring system assigns points as follows: 1
- Congestive heart failure: 1 point
- Hypertension: 1 point
- Age ≥75 years: 1 point
- Diabetes mellitus: 1 point
- Prior Stroke or TIA: 2 points
The total score ranges from 0-6 points, with higher scores indicating progressively higher annual stroke risk. 1
Risk Stratification and Annual Stroke Rates
The CHADS₂ score reliably predicts stroke risk across three categories: 1
- Low risk (score 0): 1.0% annual stroke rate
- Low-moderate risk (score 1): 1.5% annual stroke rate
- Moderate risk (score 2): 2.5% annual stroke rate
- High risk (score 3): 5.0% annual stroke rate
- Very high risk (score ≥4): ≥7% annual stroke rate
The absolute stroke risk varies 20-fold among atrial fibrillation patients based on these risk factors, making stratification essential for treatment decisions. 1
Treatment Recommendations Based on Score
Anticoagulation decisions should follow this algorithm: 1
- Score 0: Aspirin 75-325 mg daily (warfarin reduces risk by 60%, aspirin by 20% compared to placebo) 1
- Score 1: Either warfarin (INR 2-3) OR aspirin 75-325 mg daily, considering patient preferences, bleeding risk, and access to INR monitoring (NNT=100 to prevent 1 stroke per year with warfarin) 1
- Score ≥2: Warfarin (INR 2-3) is recommended 1
For patients ≥75 years, some experts recommend a lower INR target of 1.6-2.5, though this is not universally accepted. 1
Critical Clinical Considerations
Patients with prior stroke or TIA warrant special attention. All nonvalvular atrial fibrillation patients with prior stroke or TIA should be considered high risk and treated with anticoagulation regardless of their numerical CHADS₂ score. 1 Although these patients score only 2 points on the CHADS₂ scale (appearing "moderate risk"), validation studies show they actually experience 10.8 strokes per 100 patient-years—far exceeding the moderate risk category. 1
Blood pressure control is critically important in atrial fibrillation patients receiving anticoagulation, as intracerebral hemorrhage (the most devastating complication of anticoagulation) is exquisitely sensitive to blood pressure control. 1 Hypertension with systolic BP ≥160 mmHg is an independent risk factor for stroke in atrial fibrillation. 1
Validation and Predictive Performance
The CHADS₂ scheme was validated in large cohort studies and clinical trials, demonstrating superior predictive value (c-statistic 0.82,95% CI 0.80-0.84) compared to earlier stratification schemes. 1 The score has been prospectively validated across multiple independent cohorts. 1
The score also predicts stroke severity and outcomes. High-risk CHADS₂ scores predict poor neurological outcomes after atrial fibrillation-related stroke (OR 4.17 for NIHSS ≥2, OR 2.97 for mRS ≥3) and increased mortality (HR 3.01 for all-cause death). 2
Common Pitfalls and Limitations
Do not apply CHADS₂ to valvular atrial fibrillation—this score is specifically for nonvalvular atrial fibrillation only. 1
Rhythm control does not reduce stroke rates—antithrombotic therapy remains the mainstay for stroke prevention regardless of whether rate or rhythm control strategy is pursued. 1
Excellent anticoagulation control is essential to achieve the expected benefit, particularly for patients with CHADS₂ score of 1 where the NNT is 100. 1
The score has been largely superseded by CHA₂DS₂-VASc in contemporary practice, which includes additional risk factors (vascular disease, age 65-74, female sex) and demonstrates superior sensitivity for identifying truly low-risk patients. 3, 4 The CHA₂DS₂-VASc score has a broader range (0-9 vs 0-6) allowing more refined risk stratification and better discriminates stroke risk among patients with low CHADS₂ scores. 3